tion on family size, were obtained in a survey, and the data were then analyzed with the scores of verbal reasoning tests administered by the schools (the Junior Leaving, or 11-plus, Examination). Scott's findings agreed with those obtained elsewhere: both intelligence test scores and growth decreased as family size increased. In addition, he presents the data shown in Table 12, which shows "a tendency for intelligence both to increase with height and to lessen as family size increases." Scott concluded that his findings showed: (a) Height, weight and verbal reasoning scores all tend to decrease as family size increases. (b) Independently of family size, height and verbal reasoning scores tend to rise and fall in sympathy. (c) Family size is more closely associated with verbal reasoning than with height. (45, p. 169) TABLE 12 Average Verbal Reasoning, Standardized Scores (11+ Examination) by Height and Family Size, England, 1949 Number of Children in Family Height (cm.) 2 3 Average Value Reasoning Scoresa 4 and More Less than 130.0 96.2(19) 101.2(35) 97.5 (16) 94.5 (33) 130.0-134.9 102.1 (58) 100.4 (88) 101.4(52) 94.1 (55) 135.0-139.9 108.4(70) 107.0(119) 102.4 (80) 100.7 (44) 140.0-144.9 108.5 (54) 106.0(104) 109.5 (37) 99.9 (44) 145 and over 108.5 (43) 107.5 (50) 106.5 (28) 102.8(16) aFigures in parentheses are the number of children in each class. Source: Scott (45). The data in Table 12 also suggest that even if the constellation of causal factors interacting in a large family is such that the children grow well, they are still likely to suffer in their intellectural development. The data seen in Tables 5 and 9 from the British National Survey of Health and Development (35, 36) support this concept, although from a different angle. In the upper-middle families of that study the negative correlation between family size and growth eventually disappeared, but the negative correlation between family size and intelligence remained. These families apparently can and do compensate for increased family size sufficiently to support the growth of their34-35) Source: Douglas (35).he negative correlation between intelligence and family size varied among the classes. It was "clearly apparent" in children of fanners, manual laborers, and clerical workers, "negligible" in children from the managerial class, and "barely discernible" among those from the professional class.